The purpose of this proposal is to devise and evaluate methods to predict the responsiveness of human prostatic cancer to androgen ablation. Hormonal responsiveness is believed to depend on the presence of steroid receptors, but because prostatic tumors may exhibit varying degrees of heterogeneity, assays that use homogenized tissue, which masks this heterogeneity, are limited. Further progress in developing accurate methods to predict responsiveness must take heterogeneity into account in a quantitative manner. New morphological approaches are needed to provide critical insight into the relationship between steroid hormone receptors and hormonal responsiveness of prostatic cancer. We have developed a new method for 3H-steroid autoradiographic localization of androgen receptors in prostatic tissue which meets this need and which can be applied to needle biopsy specimens. We propose to use this new method to study the role of receptor measurements in predicting endocrine responsiveness of prostatic cancer. Androgen receptor-containing cells will be identified in needle biopsies of prostatic cancers using 3H-steroid autoradiography. Receptor densities will be quantified at the light microscopic level by grain counting using newly developed computer-assisted image analysis procedures. We will thus determine the nature and extent of heterogeneity of androgen receptor distribution within prostatic cancer. Quantitative aspects of androgen receptor autoradiography of specimens taken just prior to androgen ablation therapy will be correlated with quantitative aspects of clinical response in patients with metastatic prostatic cancer (stage D2) in order to determine whether these receptor measurements are of value in predicting androgen responsiveness. We will also investigate the relationship of androgen receptor content and distribution to other markers of differentiation such as prostate-specific acid phosphatase, prostate-specific antigen, progesterone receptors and estrogen receptors, using morphological techniques. We will determine whether the presence of these markers correlates with the presence of androgen receptors, and whether any of these other markers improve the predictive value of androgen receptor measurements in assessing the androgen responsiveness of prostatic cancer.